| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 37, 35620-35628, September 12, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶
From the
Radiation Biology Center, Kyoto
University, Kyoto 606-8501, Japan and the
Laboratory of Protein Informatics, Research
Center for Structural Biology, Institute for Protein Research, Osaka
University, Suita 565-0871, Japan
Received for publication, May 13, 2003 , and in revised form, June 13, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
/
domain and a C-terminal
helical domain connected by an interdomain loop. RD-2a is within this loop,
RD-1 is at the N-terminal 50 amino acids, and RD-2b at the following 31 amino
acid residues of the helical domain. Either RD-2a or RD-1 is required for
interaction with the CLOCK: BMAL1 heterodimer, and either RD-1 or RD-2b is
required for the nuclear translocation of CRY. Both of these functions are
prerequisites for the transcriptional repressor activity. The functional
nuclear localizing signal in the RD-2b region also was identified. The
sequence is well conserved among repressor-type CRYs, including mCRY1.
Mutations in the nuclear localizing signal of mCRY1 reduce the extent of its
nuclear localization. These findings show that both nuclear localization and
interaction with the CLOCK:BMAL heterodimer are essential for transcriptional
repression by CRY. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The zebrafish provides an attractive vertebrate model for biological clock analyses. Several of its clock genes have now been identified, and in vitro analyses has shown that the zebrafish negative feedback loop consists of components similar to those of mammals. zPER and zCRY act as negative regulators, and zCLOCK and zBMAL as positive elements (14–20). Four repressor-type zCRYs (zCRY1a, -1b, -2a, and -2b) have been identified in zebrafish. They repress the activities of mouse CLOCK and BMAL, as well as those of zebrafish CLOCK and BMAL, indicative that the basic function of repressor-type zCRYs is the same as that of mCRYs (16, 19, 20). A unique feature of zebrafish CRY is the presence of extra paralogous genes. In addition to repressor-type CRYs, the zebrafish has a unique CRY, zCRY3, that despite high structural similarity to repressor-type CRYs, lacks transcriptional repression (16). zCRY3 therefore can be classified as a non-repressor-type CRY, but its function has yet to be identified.
CRYs are members of the DNA photolyase/cryptochrome protein family (21–23) that comprise such functionally diverse members as DNA photolyase and CRY. DNA photolyase is a unique enzyme that repairs a UV-induced DNA damage in a light-dependent manner (24–26), and CRYs function in the circadian system. Despite functional diversity, the members of this protein family have a high degree of sequence similarity and flavin adenine dinucleotide (FAD) as a common cofactor. Although animal CRYs have a crucial role in the central circadian clock, because of structural complexity their functional domains have not been well characterized. CRYs require FAD as a cofactor to maintain proper conformation, and FAD binding sites are present within the N-terminal conserved regions (27–30). Any deletion in the N-terminal region therefore causes complete loss of the CRY function. This interferes with the deletion analysis commonly used to determine the functional domains of a protein. We took advantage of the similarity in nucleotide/amino acid sequences and difference in transcriptional repressor activities of the repressor-type zCRY (zCRY1a) and non-repressor-type zCRY (zCRY3) to form chimeras, which allowed us to map the region responsible for transcriptional repressor activity. We identified the sequence elements required for the interaction of CRY with other clock proteins, as well as for regulation of the subcellular distribution and transcriptional repression. The findings for these chimeras show that both the nuclear localization and interaction with the CLOCK:BMAL heterodimer are essential for transcriptional repression by CRY.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
|
GFP fusions were generated as follows: the SalI-EcoRI fragment bearing zCRY1a NLS (amino acids 265–282) was excised from pcDNA-zCRY1a, then ligated into the corresponding site of pGFP-CI, generating pGFP-zCRY1a (amino acids 265–282). The segment encoding amino acids 265–282 of zCRY3 was amplified by the use of oligonucleotides with SalI or EcoRI then ligated into the SalI-EcoRI site of pGFP-CI, generating pGFP-zCRY3 (amino acids 265–282).
The coding region of mouse Cry1 cDNA was obtained from mouse brain RNA by a reverse transcriptase-PCR then ligated into pcDNA-V5, generating pV5-mCRY1. Amino acid substitutions in mCRY1 NLS (amino acids 265–282) of mCRY1-V5 were introduced by a two-step PCR scheme that used primers encoding the mutated nucleotide. The PmaCI site was introduced at bp 375 of the nucleotide sequence encoding mCRY1 by a two-step PCR scheme, generating pV5-mCRY1 (PmaCI) and pV5-NLS-mutated mCRY1 (PmaCI). The HindIII-PmaCI fragment bearing zCRY3 (amino acids 1–125) was ligated into the corresponding sites of pV5-mCRY1 (PmaCI) and pV5-NLS-mutated mCRY1 (PmaCI), generating the zC3-mCRY1 chimeras (Fig. 7).
|
All the constructs described were verified by sequence analysis. Other plasmids used in this study have been described elsewhere (16, 19, 20).
Cells, Transfection, and Luciferase Assay—NIH3T3 cells were grown in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 5% calf serum. COS7 cells were maintained in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum. The day before transfection, both types of the cells were each plated in 12-well plates, and then were transfected the next day with 50 ng of firefly luciferase reporter plasmid, 20 ng of sea pansy luciferase reporter plasmid (pRL-SV40 (Promega)), and expression plasmids (indicated in each figure), by the use of LipofectAMINE Plus according to the manufacturer's instructions (Invitrogen). As reporter plasmids, the E-box element and its flanking sequences within the promoter/enhancer region of mouse vasopressin and five GAL4-binding sites were, respectively, cloned into the pGL-Basic vector (Promega) for the luciferase-reporter and two-hybrid assays (designated as mAVP-pGL and 5G-pGL). Total amounts of expression plasmids were adjusted by adding the pcDNA3.1 vector as the carrier. The preparation of cell lysates and the dual luciferase assays, using the dual-luciferase reporter assay system according to the manufacturer's instructions (Promega), were performed 24 h after transfection. Firefly and sea pansy luciferase activities were quantified by means of a luminometer, with the firefly luciferase activity normalized for transfection efficiency based on the sea pansy luciferase activity. All experiments were done three times.
Antibodies—Polyclonal antisera against the glutathione S-transferase-fused zCRY3 C-terminal polypeptide (amino acids 506–598) were raised in rabbits. Anti-VP16 antibody was purchased from Clontech, anti-FLAG from Sigma, anti-V5 from Invitrogen, and anti-GFP from Roche Diagnostics Corp.
Co-immunoprecipitation—Co-immunoprecipitation was done as previously described (7), with some modifications. COS7 cells were seeded in 6-cm dishes, and were transfected the following day with the expression plasmids described in each figure. The cells were washed twice with phosphate-buffered saline (PBS) 24 h after transfection, homogenized in binding buffer (150 mM NaCl, 5 mM EDTA, 0.5% Nonidet P-40, and 50 mM Tris-HCl, pH 7.5) containing protease inhibitor mixture tablets, and then clarified by centrifugation for 10 min at 10,000 x g. Total protein (100 µg) from the supernatant was incubated with 15 µl of protein A/G-agarose beads (Santa Cruz) for 1 h at 4 °C, after which the material was centrifuged. The supernatant was incubated for 12 h at 4 °C with either the anti-V5 mouse antibodies (Invitrogen) or the anti-VP16 rabbit antibody, and 15 µl of protein A/G-agarose beads. The beads were then washed three times with binding buffer, boiled in SDS sample buffer, and centrifuged. The supernatant was separated by SDS-PAGE and analyzed by Western blotting, as described below.
Western Blot Analysis—Total protein (10 µg), extracted from the cells as described previously, was separated by SDS-PAGE in a 6.5% gel and transferred electrophoretically onto a polyvinylidene difluoride membrane. The membrane was blocked with 7% nonfat milk and incubated with the mouse anti-FLAG antibody (Sigma), the mouse anti-V5 antibody (Invitrogen), the rabbit anti-zCRY3 antibody, or the rabbit anti-VP16 antibody for 1 h at room temperature. The blots were incubated with the appropriate secondary antibody, peroxidase-conjugated anti-mouse or anti-rabbit IgG antibody (Santa Cruz), and blots were developed with the ECL Western blotting detection system (Amersham Biosciences).
Immunofluorescence—NIH cells (3 x 105) were seeded on glass coverslips in 6-well dishes and transfected the following day (as described above) with 1 µg of total DNA per well. Thirty hours after transfection, the cells were washed twice with PBS, fixed with 4% paraformaldehyde in PBS, washed, and blocked for 30 min at 37 °C in 1% bovine serum albumin, 0.1% Triton X-100 in PBS. The anti-V5, anti-zCRY3, or anti-VP16 antibodies were diluted in 0.5% bovine serum albumin in PBS, and then incubated with the cells for 1 h at 37 °C. The cells were washed three times with 0.1% Triton X-100 in 10% PBS, and then the cells were incubated with the fluorescein isothiocyanate- (Santa Cruz) and/or the Cy3-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, Inc.) for1hat37 °C. After several washes, the cell nuclei were stained with 4',6'-diamidino-2-phenylindole, and the cells were mounted for fluorescence microscopy.
Protein Modeling—The amino acid sequences of Escherichia coli CPD photolyase, mCRY1, zCRY1a, and zCRY3 were aligned by the use of commercially available software, GENENTYX-MAC version 8.0 (Software Development Co., Ltd.). Tertiary models of mCRY1 and zCRY1a were constructed by comparative modeling based on the structure of E. coli CPD photolyase (Protein Data Bank code 1dnp [PDB] ), using our original programs: a loop search method for the backbone structure (31), a dead-end elimination method for the side chain structure (32), and a conformation energy minimization method for structure refinement (33), using the AMBER force field (34). The quality of the model structure was examined by the program PROCHECK (35), and it was shown to have the quality corresponding to a crystal structure with 2.0 to 2.5-Å resolution. The atom coordinates of the model structure are available from the corresponding author when requested.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
/ domain (residues
1–127) and a C-terminal helical one (residues 214–557), which are
connected by a long interdomain loop (residues 128–213). The first
(residues 1–125) and second (residues 126–196) regions were
designed, respectively, to carry the N-terminal / domain and
subsequent interdomain loop. The remaining 197–557 region has a helical
domain that is the FAD binding site, and its primary structure is well
conserved within this protein family. Initially, we planned to divide this
region into three parts: residues 197–293, 294–419, and
420–557, because the amino acid sequence in 294–419 is the best
conserved, whereas those in 197–293 and 420–557 have diverged to
some extent (Fig. 1A).
The region was further divided into two subregions, 197–263 and
264–293, to analyze the functionally important 197–293 region more
precisely. As shown schematically in Fig.
2A, 19 chimeras composed of reciprocal domain swaps were
generated between zCRY1a and -3.
|
Determination of Those Sequence Elements of zCRY1a Sufficient for Transcriptional Repression—Effects of the chimeric zCRYs on zCLOCK:zBMAL-mediated transcription were examined in a luciferase reporter gene assay. As reported elsewhere (16, 20), co-expression of zCRY1a efficiently inhibits zCLOCK1:zBMAL3-mediated transcription (Fig. 2B, lane 3), whereas that of zCRY3 does not (lane 4). First, two chimera series (chimeras 1–8), in which the N- or C-terminal regions of zCRY1a were replaced sequentially by the corresponding regions of zCRY3, were tested. Two chimeras, 3 with amino acids 197–557 and 6 with 1–293 of zCRY1a, maintained transcriptional repression activity (lanes 7 and 10), whereas their reciprocals, chimeras 4 and 5, lacked that activity (lanes 8 and 9). This showed that the region between amino acids 197 and 293 of zCRY1a is necessary for transcriptional repression. In fact, chimera 9, which had amino acids 197–293 of zCRY1a, had repressor activity (lane 13), whereas its reciprocal, chimera 10, did not (lane 14). To identify precisely the critical domain, the 197–293 region was divided into two subregions, 197–263 and 264–293. Four chimeras, 11, 12, 13, and 14, carrying each region, were tested for repressor activity. Chimera 11 with the 197–263 residues of zCRY1a had activity (lane 15), whereas chimera 13 with the rest of the region (264–293) lacked it (lane 17), evidence that the 197–263 region of zCRY1a is sufficient to repress zCLOCK1:zBMAL3-mediated transcription. Consistent with this conclusion, chimera 14, with all zCRY1a sequences except the 264–293 region, which was replaced by zCRY3, had repressor activity (lane 18). Unexpectedly, chimera 12, in which the 197–263 region of zCRY1a was replaced with that of zCRY3, also had repressor activity (lane 16). This suggests that besides the 197–263 region of zCRY1a sufficient for repressor activity, a second region also has activity. This second region must combine the separate regions because the two series of chimeras (chimeras 1–8), in which the N- or C-terminal regions of zCRY1a were replaced sequentially by the equivalent regions of zCRY3, showed no evidence of the presence a second region. Furthermore, one of the separate regions must be the 264–293 region of zCRY1a because its presence in chimera 12 is the only difference between chimeras 10 and 12; the former having lost activity, the latter retaining it. We therefore generated three additional chimeras, in which the 264–293 region of zCRY1a was combined with one of the remaining zCRY1a regions: 1–125 (chimera 15), 126–196 (chimera 16), both 1–125 and 294–557 (chimera 17). Of these three chimeras, 16 alone had repressor activity. Chimera 18, with only the 126–196 region of zCRY1a, had no activity (lane 22). The second region that confers repressor activity on zCRY3 therefore is a combination of the 126–196 and 264–293 regions.
The expression levels of the chimeras that lacked transcriptional repressor activity were examined. After transfection into NIH3T3 cells of the expression vectors that encoded each chimera, the cell lysates were analyzed by Western blotting. As shown in Fig. 2C, all the chimeras lacking repressor activity were expressed at a level comparable with that of zCRY1a, or chimera 9, a potent transcriptional repressor. This excludes the possibility that attenuation of transcriptional repressor activity in the chimeras was because of protein instability.
Two regions in zCRY1a that are responsible for the repression of CLOCK:BMAL-mediated transcription were identified: amino acids 197–263 of zCRY1a, and a combination of residues 126–196 and 264–293 of zCRY1a. Each of these regions is sufficient for repressor activity because each alone has independent activity. For simplicity, the respective regions 197–293, 126–196, and 264–293 hereafter are designated RD-1, RD-2a, and RD-2b (Fig. 1, A and C).
Determination of the Sequence Elements of zCRY1a Required for Its Interaction with zCLOCK1 or zBMAL3—For the repression of CLOCK:BMAL-mediated transcription, CRY association with CLOCK and BMAL is important (8, 19, 36). Recently, we showed that zCRY1a associates with both the zCLOCK1 and zBMAL3 proteins, whereas zCRY3 does not (20). We therefore used chimeras to examine whether the three regions RD-1, RD-2a, and RD-2b are sites of interaction with CLOCK and BMAL. The ability of each chimera to interact with zCLOCK1 and zBMAL3 was tested (Fig. 3A).
First a mammalian two-hybrid assay was used (Fig. 3B), in which zCRY fused to the GAL4 DNA-binding domain (GAL4) was co-expressed with zCLOCK1 or zBMAL3 fused to the VP16 transactivation domain (VP16) in NIH3T3 cells. If GAL4-zCRY interacts functionally with VP16-fused protein, VP16 would be recruited to the vicinity of the promoter and cause transactivation. Four chimeras, 11, 13, 16, and 18, which bear the RD-1 or RD-2 regions, were tested. When co-expressed with VP16-zCLOCK1 or VP16-zBMAL3, chimera 11 with the 196–263-(RD-1) region of zCRY1a caused transactivation (Fig. 3B, lanes 4 and 15), indicative that the 196–263-(RD-1) region of zCRY1a is sufficient for association with the CLOCK:BMAL heterodimer. Chimera 16 with both RD-2a and RD-2b also caused transactivation (lanes 8 and 19), whereas chimeras 18 and 13 with RD-2a or RD-2b did not (lanes 6, 9, 17, and 20). The presence of both RD-2a and RD-2b in CRY therefore is necessary for interaction with CLOCK and BMAL, as well as for transcriptional repression activity.
In the nucleus, CLOCK and BMAL form a heterodimer that functions as a transcriptional activator, therefore CRY must enter the nucleus to interact with the heterodimer and subsequently repress its activity (7, 36). One reason why two distinct regions are needed for transcriptional repression could be that one is responsible for the nuclear localization of CRY and the other for direct interaction with the heterodimer. To examine this probability, interactions also were investigated by immunoprecipitation analysis (Fig. 3C). To exclude the possibility that interaction is hard to detect because of different subcellular localizations of the proteins, two kinds of cell lysates that expressed different proteins were mixed in vitro and incubated to form a complex, after which immunoprecipitation assays were done. Two types of COS7 cell extracts were prepared; one derived from COS7 cells transfected with plasmids encoding FLAG-zCLOCK1 and zBMAL3-V5, the other from cells transfected with VP16-zCRY1a, zCRY3, or a chimera. These extracts first were mixed, incubated on ice, then underwent immunoprecipitation by the anti-FLAG antibody. The proteins precipitated were examined by Western blotting. Chimera 18 with the RD-2a region of zCRY1a and chimera 11 with the RD-1 region co-immunoprecipitated with zCLOCK1-zBMAL3 (Fig. 3C, lanes 2 and 3), whereas chimera 17 with the RD-2b residues of zCRY1a, did not (Fig. 3C, lane 5). These immunoprecipitation results indicate that the RD-2a and RD-1 regions of zCRY1a are the sites of interaction with the zCLOCK1-zB-MAL3 heterodimer, and that either region is sufficient for that interaction. They also suggest that the RD-2b region, an element in the second region, must have some other function. For the two-hybrid assay, the interacting proteins must be present together in the nucleus to transactivate the reporter gene. For the immunoprecipitation assay used in this study, however, the proteins do not need to be co-localized. The results presented in Fig. 3C therefore indicate that the sequence elements responsible for nuclear localization are present in the RD-1 and RD-2b regions of zCRY1a. Whether these regions have the ability to localize the zCRY protein in the nucleus therefore was examined.
Determination of the zCRY1a Regions That Regulate Subcellular Distribution—Elsewhere, we reported that zCRY1a is located in the nucleus, whereas zCRY3 mainly is distributed in the cytoplasm (20). The chimeras therefore also can be used to map the regions responsible for subcellular localization. VP16-tagged or non-tagged chimeric proteins were expressed in NIH3T3 cells, and their cellular locations were determined by immunofluorescence (Fig. 4). Subcellular locations were the cytoplasm (N < C), both the cytoplasm and nucleus (n = C), or the nucleus (N > C).
|
Four chimeras, 11, 13, 16, and 18, which bear one or two of the RD-1, RD-2a, and/or RD-2b regions, were tested for cellular localization (Fig. 4C, lanes 7–18). Except for chimera 18, they were mainly in the nucleus. This is consistent with our prediction that the presence of either the RD-1 or RD-2b regions of zCRY1a are sufficient for the nuclear localization of zCRY and that the RD-2a region lacks such activity.
Identification of a Functional NLS within the Repressor-type of CRY—Proteins larger than 48 kDa require a specific sequence, the nuclear localizing signal (NLS), to be targeted to the nucleus (37). The RD-1 and RD-2b regions are responsible for the nuclear localization of zCRYs. Our search for the consensus NLS sequence in zCRY1a found an NLS-like sequence (residues 265–282) in the RD-2b region. This sequence is highly conserved in repressor-type CRYs from different species but varies substantially in zCRY3 (Fig. 5A). pGFP-zCRY-NLS constructs were generated to determine whether the sequence functions in the nuclear import of zCRY. Amino acids 265–282 of zCRY1a or zCRY3 were inserted into the C-terminal end of GFP, which, respectively, generated pGFP-zCRY1a-(265–282) or pGFP-zCRY3-(265–282). Their subcellular localizations then were determined after transfection into NIH3T3 cells (Fig. 5B). GFP-zCRY1a-(265–282) mainly was limited to the nucleus, whereas the control GFP was present in both the cytoplasm and nucleus (Fig. 5, B and C, lanes 1–6). Unlike GFP-zCRY1a-(265–282), GFP-zCRY3-(265–282) was mainly distributed in both the cytoplasm and nucleus (Fig. 5, B and C, lanes 7–9). These findings indicate that residues 265–282 of zCRY1a constitute a functional NLS and that the distinct subcellular localizations in zCRY1a and -3 depend in part on different NLS activities.
|
In the RD-1 region, there was no consensus NLS sequence. The GFP-RD-1 fusion protein (pGFP-zCRY1a-(197–263)) was mainly restricted to the nucleus (data not shown). The RD-1 region therefore carries an unknown NLS.
The NLS-(265–282) identified in zCRY1a is well conserved among the repressor-type CRYs of various organisms (Fig. 5A). To determine whether the identified sequences also function as an NLS in other repressor-type CRYs, we determined whether disruption of the NLS of mouse CRY (mCRY) affects its subcellular distribution. Two types of NLS-mutated mCRY1s (mut-2 and mut-9) were generated by replacing two or nine amino acids of the NLS with the corresponding amino acids of zCRY3 (Fig. 6A). Both mutant proteins were expressed at levels comparable with the level of wild-type mCRY1 (Fig. 6B), indicative that the two types of substitutions did not affect the stability of the mCRY1 protein.
|
The subcellular distribution of each mCRY1 mutant was investigated (Fig. 6, C and D). When expressed in NIH3T3 cells, wild-type mCRY1 mainly was detected in the nucleus. Interestingly, both types of mutations in the NLS produced cytoplasmic distribution of mCRY1 (Fig. 6, C and D, lanes 4–9), evidence that the intact NLS (residues 265–282) of mCRY1 is necessary for its nuclear localization.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Thus far, crystal structures have been determined for four members of the
DNA photolyase/Cryptochrome family; three CPD photolyases, one each from
E. coli, Anacystis nidurans, and Thermus thermophilus
(39–41),
and one Cryptochrome from Synechosistis sp.
(42). These proteins have
similar three-dimensional structures. They are folded in two domains, an
/ and a helical domain connected by a long interdomain loop. The
helical domain has all the residues that bind FAD. FAD binds noncovalently at
the bottom of the cavity formed between the two distinct lobes of that domain.
As stated in the description of the strategy for determining the domain to be
exchanged, the RD-2a region constitutes the major part of the interdomain
loop, RD-1 and RD-2b covering the N-terminal third of the helical domain
(Figs. 1C and
7). The amino acid residues
that are well conserved in all the repressor-type CRYs, but not in
non-repressor-type CRY (zCRY3), are the most promising candidates for the
critical site of each function. These residues are Ser-129, Leu-135, Gly-143,
Tyr-150, Ser-158, and Met-160 in RD-2a, and Leu-205, Ala-208, Met-239,
Asn-240, Ala-241, Asn-242, and Ala-246 in RD-1. These repressor type-specific
amino acid residues and the NLS sequences in RD-2b were mapped on a modeled
structure of mCRY1 (Fig. 7). The entire RD-2b NLS region and some of the repressor type-specific residues
of RD-1 and RD-2a are exposed to the solvent. Interestingly, they are
clustered in a limited area, in which the RD-2b NLS region is at the center,
the regions with the repressor type-specific residues of RD-1 and RD-2a being
positioned on each side. This clustering may explain why two distinct regions
were identifiable as CLOCK-BMAL1 interaction sites. RD-1 and RD-2a constitute
one CLOCK-BMAL1 interaction domain, but interaction with either RD-1 or RD-2a
might be strong enough to trigger transcriptional repressor activity.
Interestingly, the cluster is just at the rim of the FAD-containing cavity.
FAD is essential in the repair reaction of DNA photolyase
(22,
39,
42) that occurs by electron
transfer from light-excited FADH–to the UV-induced DNA lesion. This
reaction is initiated by entrance of the damaged site into the hole, providing
access to the FAD bound at the bottom of the cavity. The structure of the
cavity and position of FAD in it therefore constitute the area essential for
the repair reaction of DNA photolyase. The role of the FAD in the cavity on
the function of CRY, however, is not clear. The position of the region
responsible for transcriptional repressor activity at the rim of the cavity
suggests a regulatory role for FAD in transcriptional repressor activity.
CRY is thought to act as a transcriptional repressor by interacting directly with the CLOCK:BMAL heterodimer in the nucleus (7, 8, 19, 38); and both interactions with the heterodimer and nuclear translocation seem to be essential for the transcriptional repressor activity of CRY. Our findings provide several lines of evidence that support this. First, the area responsible for nuclear localization and for interaction with CLOCK and BMAL1 are separable into two distinct regions: the former is RD-2b, the latter is RD-2a. Transcriptional repressor activity occurred only when both regions were present in the molecule. Second, there was good correlation between the transcriptional repressor activity and nuclear localization ability in several types of chimeric or mutated CRYs. We identified a functional CRY NLS (residues 265–282) in zCRY1a and mCRY1 that is highly conserved in repressor-type CRYs of different organisms. Mutations in the NLS of mCRY shift subcellular localization from the nucleus to the cytoplasm and decrease its ability to repress CLOCK:BMAL-induced transactivation (data not shown). Taken together, these findings clearly indicate that both nuclear localization and interaction with the CLOCK:BMAL1 heterodimer are prerequisites for the transcriptional repressor activity of CRY.
Initially, we speculated that mCRY1 and zCRY1a have identical structures and functions. This was based on two facts: the high homology of the primary structures of both CRYs and the in vitro abilities of these CRYs to repress the transcription activated by mouse- or zebrafish-derived CLOCK and BMAL1 (16, 20). The present findings are consistent with this speculation. Except for one amino acid in each region (D203E and K274R) (Fig. 1A), the amino acid sequences in the RD-1 and RD-2b regions are identical. Furthermore, the RD-2b region functions as a nuclear localizing domain in mCRY1 (Fig. 6). The amino acid sequences of the RD-2a region vary to some extent, but the sequences of the N-terminal halves of RD-2a, in which the repressor type-specific amino acid residues (Ser-129, Leu-135, Gly-143, Tyr-150, Ser-158, and Met-160) are clustered, are well conserved in the two CRYs (Fig. 1A). These observations confirm our speculation. We also tried to determine the interaction site at the amino acid level in RD-1 and RD-2a, but were unsuccessful because no point mutation at any of the repressor type-specific amino acid residues had a clear effect on the repressor activity of chimeric zCRYs (data not shown). Not just one, but several amino acid residues are responsible for that activity. The introduction of clustered mutations at these residues will help us to determine which are the critical amino acids in these regions and providing a clearer understanding of the regulatory mechanism of this fascinating CRY protein.
|
FOOTNOTES |
|---|
¶ To whom correspondence should be addressed: Radiation Biology Center, Kyoto University, Yoshidakonoe-cho, Sakyo-ku, Kyoto 606-8501, Japan. Tel.: 81-75-753-7555; Fax: 81-75-753-7564; E-mail: todo{at}house.rbc.kyoto-u.ac.jp.
1 The abbreviations used are: TIM, TIMELESS; CRY, Cryptochrome; PER, PERIOD;
GFP, green fluorescent protein; PBS, phosphate-buffered saline; NLS, nuclear
localization signal; z, zebrafish; m, mouse.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Hirayama, S. Cho, and P. Sassone-Corsi Circadian control by the reduction/oxidation pathway: Catalase represses light-dependent clock gene expression in the zebrafish PNAS, October 2, 2007; 104(40): 15747 - 15752. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Yu, D. Shalitin, X. Liu, M. Maymon, J. Klejnot, H. Yang, J. Lopez, X. Zhao, K. T. Bendehakkalu, and C. Lin Derepression of the NC80 motif is critical for the photoactivation of Arabidopsis CRY2 PNAS, April 24, 2007; 104(17): 7289 - 7294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Chaves, K. Yagita, S. Barnhoorn, H. Okamura, G. T. J. van der Horst, and F. Tamanini Functional Evolution of the Photolyase/Cryptochrome Protein Family: Importance of the C Terminus of Mammalian CRY1 for Circadian Core Oscillator Performance. Mol. Cell. Biol., March 1, 2006; 26(5): 1743 - 1753. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Hirayama, L. Cardone, M. Doi, and P. Sassone-Corsi Common pathways in circadian and cell cycle clocks: Light-dependent activation of Fos/AP-1 in zebrafish controls CRY-1a and WEE-1 PNAS, July 19, 2005; 102(29): 10194 - 10199. |
